This project focuses on developing new statistical methods, and applying new and existing statistical techniques, to analyze data from laboratory animal studies. Much of our research dealt with tumor incidence, which is the rate at which new tumors arise. We investigated methods that handle complex data structures without relying on unrealistic assumptions. An analysis was developed which permits simultaneous inference about tumors at multiple sites while accounting for within-animal correlations among tumor onset times.